A Photo Junction Field-Effect Transistor. (photojfet) Based on a Colloidal Quantum Dot. Absorber/Channel Layer

Transcription

1 SUPPORTING INFORMATION A Photo Junction Field-Effect Transistor (photojfet) Based on a Colloidal Quantum Dot Absorber/Channel Layer Valerio Adinolfi ɫ, Illan J. Kramer ɫ, Andre J. Labelle ɫ, Brandon R. Sutherland ɫ, Edward H. Sargent ɫ* ɫ Department of Electrical Engineering and Computer Science, University of Toronto, 10 King s College Road, Toronto, Ontario M5S 3G4, Canada Index Figure S1: PbS Colloidal Quantum Dot absorbance... 2 Figure S2: Analytical Model... 3 Figure S3: Fall and rise time... 4 Figure S4: Rectification... 5 Figure S5: Responsivity... 6 Figure S6: Spectral gain... 6 Figure S7: Carrier lifetime, mobility and gain estimation... 7 Figure S8: Detectivity

2 Figure S9: Noise current... 9 Figure S1: PbS Colloidal Quantum Dot absorbance Absorbance (a.u.) Wavelength (nm) Figure S1: Absorbance spectrum of PbS CQDs. The plot exhibits the characteristic excitonic peak close to 950nm. 2

3 Figure S2: Analytical Model The analytical relation between gain, dark current density and frequency has been calculated according to the following model. The photoconductive gain can be written as: (1) = / Where is the charge lifetime, and is the transit time of the carriers. The latter quantity can be expressed using the following equation: (2) = / Where is the channel length, V the applied voltage, and the carrier mobility. Since determines the response time for a given photoconductive material, the maximum operative frequency can be expressed as: (3) =1/ If we now use the following definition for the current density: (4) = / Where is the electron charge, and is the carrier density, and we combine the equations together, we finally obtain the relation: (5) =. We plotted equation (5) by using the software MATLAB. We used the following numerical values: photojfet: =5 10, =5 photoconductor: =1 10, =5 The difference in the carrier density reflects the presence of the depleted channel in the photojfet. The channel length has been chosen equal for both the devices in order to allow for a direct comparison. 3

4 Figure S3: Fall and rise time Figure S3: Fall and rise time are reported. The values are extracted by measuring the time taken by the signal to change from 10% to 90% of the step height. The response time is substantially identical confirming there is only one time constant dominating the system. 4

5 Figure S4: Rectification Figure S10: The MoO 3 /CQD rectifying IV curve is shown for different thickness of the CQD layer (150, 300 and 450 nm) 5

6 Figure S5: Responsivity Figure S5: Responsivity of the photojfet as a function of the incident optical power. Three curves are shown for different values of the drain source voltage V DS. Figure S6: Spectral gain 6

7 Figure S6: Spectral gain (EQE) of the photojfet. The measurement was performed at a constant incident power (632 pw). The plot resembles the CQD absorption (see Figure S1) suggesting good extraction of carriers for all the wavelengths of interest. A small decrease of the gain at short frequency is attributable to defect states at the CQD/substrate interface. Figure S7: Carrier lifetime, mobility and gain estimation a) b) Figure S6: a) Carrier lifetime for two different samples of TBAI treated CQD film. b) FET measurements for a TBAI treated CQD film The photoconductive gain can be calculated as: = 7

8 Where is the quantum efficiency (or absorbance), the transit time and the minority carrier lifetime. For a channel length =2.5, an applied voltage =30 and a measured mobility = /, we can calculated the transit time = = Figure S6 a) shows a carrier lifetime ~ 5 10 that, as expected, well matches with the response time of the device. Using a measured absorbance ~80%, at =450 we find for the gain a value of: ~ This value is consistent with the experiments that show a gain of 10 at a =30. The slightly lower experimental result can be ascribed to parasitic series resistances not considered in this model. Figure S8: Detectivity Figure S8: Detectivity (Hz 1/2 cm/w, Jones) as a function of the wavelength. The data have been collected at a constant incident optical power of 632 pw, applying a drain source voltage V DS = 5 V. 8

9 Figure S9: Noise current Figure S9: noise current is measured as a function of the dark current. As expected the device is dominated from the dark current shot noise (solid red line). 9

10 10